zim-wiki team mailing list archive

[Jaap Karssenberg <jaap.karssenberg@xxxxxxxxx>]

Hi Sara,

Yes, we need to improve on release testing, as I just wrote in a new mail
thread, I think we should start working with release candidates for future
releases. However, the problem remains that putting out a release candidate
with a lot of issues still impacts a lot of people who may want to test,
but also do want to do actual work. I fear that not pushing the candidates
to the PPA will mean they are only seen by a limited audience.

Best way to improve the quality before anyone is impacted is by improving
the automated test suite. I was very surprised that the issue with the
headers going missing was not caught by the test suite. Part of my fix is
to make sure the test will now detect this if it ever happens again.

Maybe for other parts of your workflow you can put small scenarios in a bug
report (yes please use github issues) that can be turned in an automatic
test. That will better secure bugs not returning without the burden of
manual testing.

Regards,

Jaap



On Mon, May 1, 2017 at 12:19 PM Sara Ziner <sara.ziner@xxxxxxxxxxxxxx>
wrote:

> Hi Jaap,
>
> thanks a lot for the kind answer!
> Don't get me wrong: With all these little this-and-that's, Zim is simply
> awsome.
> Let alone the verbatim storage in text files and the serverless setup. In
> fact these both are the reason why I have chosen to use Zim strategically
> for each and every note I take, image set I want to comment and archive,
> even to keep my personal document file: There is no lock-in to some strange
> data format which tomorrow can cease to exist.
> I even restrain from upgrading my corporate MacBook to Sierra because I
> heard there are issues with Zim.
>
> Regarding the testing:
> Now that zim is maintained in GitHub, I would happliy vounteer to do a
> manual test with my personal workflow before new releases. You even could
> employ a dev branch that can be merged to produce a release to support
> close test cycles.
> I admit that my workflow is a bit biased towards image archiving (off the
> Notebook's folder to allow SCM for that one) and automatic generation,
> manipulation and tree movement of pages from scraped web content – but
> maybe this will complement other tester's workflows.
>
> (And, regarding the background spawning, I had to learn that I now at
> least see quite error messages and would be able to report them. Do you
> plan to switch to GitHub issues for error messages? – Would be fantastic
> because they work like a charm.)
>
> Again: Thanks for caring!
> Sara
>
>
>
>
>
> 2017-05-01 8:52 GMT+02:00 Jaap Karssenberg <jaap.karssenberg@xxxxxxxxx>:
>
>>
>> Hi Sara,
>>
>>
>> _______________________________________________
> Mailing list: https://launchpad.net/~zim-wiki
> Post to     : zim-wiki@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~zim-wiki
> More help   : https://help.launchpad.net/ListHelp
>